summaryrefslogtreecommitdiff
path: root/fs/io_uring.c
diff options
context:
space:
mode:
authorAlmog Khaikin <almogkh@gmail.com>2022-03-21 12:00:59 +0300
committerJens Axboe <axboe@kernel.dk>2022-03-21 15:33:29 +0300
commit649bb75d19c93f5459f450191953dff4825fda3e (patch)
tree6fbdce114ff684f42607f4145e086bbb9a2d3fa9 /fs/io_uring.c
parentf63cf5192fe3418ad5ae1a4412eba5694b145f79 (diff)
downloadlinux-649bb75d19c93f5459f450191953dff4825fda3e.tar.xz
io_uring: fix memory ordering when SQPOLL thread goes to sleep
Without a full memory barrier between the store to the flags and the load of the SQ tail the two operations can be reordered and this can lead to a situation where the SQPOLL thread goes to sleep while the application writes to the SQ tail and doesn't see the wakeup flag. This memory barrier pairs with a full memory barrier in the application between its store to the SQ tail and its load of the flags. Signed-off-by: Almog Khaikin <almogkh@gmail.com> Link: https://lore.kernel.org/r/20220321090059.46313-1-almogkh@gmail.com Signed-off-by: Jens Axboe <axboe@kernel.dk>
Diffstat (limited to 'fs/io_uring.c')
-rw-r--r--fs/io_uring.c7
1 files changed, 7 insertions, 0 deletions
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 1a65d7880440..48f4540d7dd5 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -8042,6 +8042,13 @@ static int io_sq_thread(void *data)
needs_sched = false;
break;
}
+
+ /*
+ * Ensure the store of the wakeup flag is not
+ * reordered with the load of the SQ tail
+ */
+ smp_mb();
+
if (io_sqring_entries(ctx)) {
needs_sched = false;
break;